I was recently reading an interesting post on Emily Short’s blog about how she handled conversation in her most recent Interactive Fiction project. You can read the post in question here.
The gist of it, for those who’d prefer a synopsis, is that she’s moved away from the conventional model, in which the player chooses something to say, and the NPC responds, then the player chooses something else to say (possibly with new options now), and the NPC responds again, and so on. Instead, she’s given the NPC as much agency in the conversation as the player – the NPC will now have a list of things he or she wishes to say, and an order to say them in, and will go down the list until they’ve said everything they want. In the meantime, the player can say or do whatever he or she likes, and these words and actions will tend to add and/or remove elements from the NPCs list.
These elements can be either immediate or postponed – which determines whether they’re added to the top or bottom of the list – and either obligatory or optional – which determines whether they’re removed from the list or not when the topic of conversation shifts.
Reading this, it occurred to me that you could use such a system to create a highly non-linear RPG, by simply extending it to cover actions as well as words, and also to include entities other than just NPCs (e.g. the environment, a whole group of monsters or even a distant kingdom could be an entity with a list of things it “wants” to do). Also, these actions would need to have a setup and a recovery time, as global events in an RPG don’t fall as neatly into a turn-based structure as conversation in a piece of IF. Aside from these changes, however, the basic idea remains unchanged.
Today, I’ve spent a lot of time thinking about how one would go about designing such a game, and I think the answer is to take the player out of the game to begin with, and come up with a story that would take place without any player intervention at all. Once you have that worked out, you can break it into one or more cause-and-effect chains, in which actions taken by one entity add actions to other entities’ to-do lists, and so on down the line. The player’s actions could then prevent some of these events from taking place, or change their outcome, or cause entities to take other actions entirely. In this way, there’s guaranteed to be a story of some sort going on, and the player can meddle in it however much he or she likes.
I think this system would be best suited to small games, because the system gets very complex very quickly, and the non-linear nature of the gameplay would give the game a high degree of replayability, so a long play-time might be more of a drawback than a positive. To manage the complexity in a larger game, the thing to do would probably be to divide the game world up into a number of isolated “islands,” (e.g. towns) whose entities only interact with others within their own island, and only major events in one place affect what happens elsewhere.
As an example, here’s an idea for a very small game I came up with, consisting of a very small number of locations and NPCs.
There are two towns. The one in which the player starts, we’ll call Hometown, and the other we’ll call Awaytown. Both towns have a few important NPCs, and a number of generic NPCs who only share rumors with the player, or give him minor tasks with minor rewards, and little impact on the main plot.
There are two ways to get from one to the other – a pass through the Mountains, and a Road through the Woods. The mountains also contain a wizard’s hut and a cave full of faeries, while the woods contain a bandit stronghold.
The major NPCs:
Hometown Sheriff, who is secretly corrupt and in cahoots with the Stablemaster (below)
Bartender, who needs a cart wheel, and knows various rumors e.g. about the bandits
Bartender’s son, who is in his late teens
Goldsmith, who makes jewelry, but is currently out of gold, and expecting a shipment soon
Awaytown Sheriff, who is a good guy, and wants to take down the bandits
Stablemaster, who is in contact with the bandits and supplies them with horses and information
Cartwright, who has a young son
Bandit Lord, who stays in the stronghold most of the time
Bandit Prince, who is the Bandit Lord’s son, and leads the group of bandits who attack the player and/or any NPCs who attempt to take the road between the towns.
Wizard, who has a powerful magic amulet
Faerie Lord, who is dying
Faerie Council, who need the magic amulet and a human child for a ritual to create a new Faerie Lord to replace their dying leader.
The Plot, without player intervention:
And so, we see that the story has a sad ending if the player does nothing; most of the nice people end up dead, and the bad guys come out ahead. Fortunately, the links in that chain are quite fragile, and the player can bend or break any one of them, and radically change the flow of events.
Early in the story, if the player talks to Bartender or Hometown Sheriff, he could be tasked with going to Awaytown to deliver the letter and/or acquire the cart wheel. If he goes to get the wheel, but doesn’t bring the letter, eventually Hometown Sheriff will find someone else to deliver it, but it will slow down the advancement of that part of the story. Meanwhile, if he does take the letter, he could deliver it as promised, advancing the story normally, or he could open it and read it. If he does so, he could expose Hometown Sheriff and raise a mob to lynch him, or he take the evil route and blackmail Hometown Sheriff into letting him in on the action. Alternately, he could take the letter to Awaytown Sheriff, who will propose that he go back to Hometown, steal Hometown Sheriff’s seal, reseal the letter, and deliver it… so that now, with the player’s assistance, Awaytown Sheriff can arrange a sting to catch the bandits – player, Awaytown Sheriff and a small posse meet up with the gold caravan and hide in the wagons to surprise the bandits.
Another way the player could find out about the Stablemaster/Hometown Sheriff conspiracy is if he happens to be out and about at nighttime in Awaytown, when the bandits come to meet with Stablemaster. The player might end up seeing them and eavesdropping.
Meanwhile, if the player chooses to travel via the mountain pass, he may run into the Faeries and/or the Wizard, depending on how much exploring he does. If he meets the Wizard, he’ll find out about the stolen Amulet – if he manages to figure out that the Faeries have it, and get it back, he’ll have the Wizard on his side. If he meets the Faeries, he may find out about their dying leader. If he’s particularly clever in conversation, he may even get some information about the ritual they have planned. The alternative to creating a new Faerie Lord is, of course, to save the current one, which requires some rare herbs from the Woods. If the player gets these in time, he can avert the abduction of Cartwright’s son. The Faeries may also reward him by giving him the amulet, which they no longer need. He could then keep it (and risk the Wizard’s wrath, if he sees him with it), or return it to the Wizard.
If instead, the player takes the road, he’ll be attacked by a small group of bandits. Assuming he wins the fight, Bandit Prince will attempt to flee. If captured, he will attempt to bargain with the player, saying he’s impressed with the player’s combat prowess, and inviting the player to join them. This would kick off a whole “evil” plotline, in which the player is with the bandits. Alternately, the player could turn Bandit Prince over to either Sheriff. Hometown Sheriff will, of course, set things up so Bandit Prince can escape, which will possibly help the player realize what’s going on. Awaytown Sheriff, on the other hand, will interrogate Bandit Prince to find out the location of the bandit stronghold, possibly leading to an attack on it, possibly with the player’s assistance… and, possibly even with the help of the Wizard and/or the Faeries, if he’s achieved happy resolution on that side of things.
Of course, the Bandit Prince might also escape, only to show up again later. Or the player might kill him, in which case the Bandit Lord, outraged over his son’s death, would seek revenge – either against the player specifically, or by attacking one or both towns, if he doesn’t have a way of determining who killed him.
Meanwhile, if the Faerie Lord does die, then the player might go to try to rescue Cartwright’s son himself, or he might join the party that ends up getting caught in the avalanche, or he might go up afterwards, to rescue them after they’re caught in the avalanche.
If this all sounds exceptionally complicated, that’s just because it’s nonlinear. There are any number of other possible interactions that could take place too – I’ve just tried to cover the main possibilities, but there are any number of ways the player character could involve himself in everyone’s business, either for altruistic reasons, personal profit, or loyalty to a specific NPC that the player has decided that he likes. Whatever happens, the overall story will probably be fairly short, but dense – this isn’t an epic game, by any means. However, the player should be able to play it several times, with a radically different experience each time through… almost any NPC or group of NPCs can end up being an ally or an enemy, depending on what the player does, and certain locations (the inside of the faerie cave and the bandit stronghold) might not even be seen at all in a given game!
Also, the possible permutations and combinations of PC-NPC and NPC-NPC interaction grow exponentially as more are added. With just four main areas and a handful of entities, the game is already quite complex – just a couple more locations to visit and a few more active parties with their own agendas, and the story would be truly unpredictable and surprising. The ultimate goal, of course, is to make an RPG that is actually a role-playing game in the true sense of the term, similar to the experience one has playing a pencil-and-paper RPG.
Gameplay: 7/10 Graphics: 7/10 Sound: 8/10 Originality: 7/10 Overall: 7/10
Play at: http://www.kongregate.com/games/TerryCavanagh/dont-look-back
Retro, in the sense of ancient history
Modern fantasy games draw plenty of elements from Greek mythology, but usually these consist of places, people, magical items and monsters, taken out of the context of their original stories. Here we have a game that takes its (minimal) storyline – and its most original gameplay feature – from the story of Orpheus. In the myth, Orpheus’ wife Eurydice dies, and he travels to the underworld to persuade Hades and Persephone to let him bring her back. They agree, but on the condition that he not look back at her until they’ve both returned to the surface. Unfortunately, in his anxiety, he can’t help looking back as they approach the surface, and thus loses her forever.
In Terry Cavanagh’s Don’t Look Back, you take the role of Orpheus, albeit anachronistically armed with a pistol, rather than a lyre. Running, jumping and gunning your way through hell, you eventually confront Hades and “persuade” him – via several bullets to the head – to liberate your dead bride. At this point, the game’s title – and the reference to the myth – come to light, as you must now return to the surface, but while moving only in a single direction; turning around even for an instant results in her disappearance, which is equivalent, in game terms, to your own death.
There are several good things about this game, and some not so good. In terms of gameplay, the level of challenge is tough but fair, and the level of frustration is kept manageable by the fact that you have unlimited lives and only need to restart the current screen when you die (or accidentally look back at your fair maiden). The difficulty curve is rather inconsistent, unfortunately. Some early screens are quite tough, and some later ones are trivial; I would have preferred to see more thought put into introducing challenges gradually. The two bosses – Cerberus and Hades – are likewise reasonably satisfying to fight, but poorly balanced in terms of difficulty; you would expect the main boss (Hades) to be the more difficult of the two, but it took me far more attempts to get past Cerberus.
Part of the problem is with the “don’t look back” mechanic. It’s original, but greatly limits the level design. Since you can’t turn or move backwards at all, the challenges actually become more simplistic, as the controls have been reduced to simply moving forward and/or jumping. Although interesting from an artistic perspective, it fails from a game design perspective for this reason. Much more interesting would have been if Cavanagh had Eurydice follow a little ways behind Orpheus, and allowed him to turn around so long as she was not in sight – for instance, if the player could get her temporarily stuck behind a wall. This would allow for much more interesting puzzles than the game currently offers.
There are likewise good and bad things to be said about the graphics; certainly, the blocky pixels and limited palette will be nostalgic for those of us who grew up on the Atari 2600 and similar machines. The palette itself is nicely-chosen as well, with everything portrayed in suitably bleak reds and burgundies. The special effects are also nice, including the rain in the outdoor scenes, and Eurydice vanishing in a puff of smoke if the player does look back. However, as is often the case with such games, one often gets the impression that the graphical “style” is often just a convenient excuse for artistic laziness – for instance, it would not be hard for a marginally competent artist to produce better-looking trees than those in the screenshot above, even with a four-color palette and low resolution.
The sound, meanwhile, is one of the game’s better features. The music suits the mood well, and I like that it comes and goes, rather than looping continuously. Eurydice’s wail when she goes up in smoke is also well done. The death sound is extremely retro.
I like the fact that this game manages to bridge the gap between being a game and an interactive art piece. Most art games feature minimal gameplay, and often no challenge whatsoever. Don’t Look Back, on the other hand, has several difficult parts, and manages to be fun, albeit simple. The artistic expression in the game is limited to the concept of the title and corresponding gameplay feature, the general graphical and audio style, and the game’s beginning and ending scenes. The rest of the time, the game focusses on being a game, as it should. All too often, those seeking to gain acceptance for games as a valid art-form try to do so by making their games less game-like; this, in my opinion, is self-defeating.
A masterpiece, this is not, but it’s still an interesting experiment. Moreover, it’s short, reasonably entertaining, and doesn’t even require a download – I would say it’s well worth investing 15-30 minutes of one’s time to give it a play through.
Anyone with any sort of creative or artistic talent has probably been approached at some time or another by a friend, family member, or random person on the Internet, asking them if they’d like to contribute to a project in some way. Usually, the people making these solicitations are looking for something to be done for free. Of course, it’s normal to want to help friends out, and especially when one is starting out, pro bono work seems like a good idea in theory. Among other things, it’s a way to get exposure, a portfolio piece, and some experience contributing to a larger endeavor instead of doing individual, personal pieces without any of the constraints present when working for a client.
However, every creative professional I’ve spoken to, in pretty much every field, has the same thing to say, which is that pro bono “clients” are by far the most troublesome. Usually, they’re more trouble than they’re worth. This is counter-intuitive, as you’d expect that most people would understand that beggars can’t be choosers, and that they’d be grateful to get anything at all from you, since they aren’t offering anything in return. I’ve often wondered why they’re almost universally so difficult to please, and I think, after my latest such experience, I’ve realized why it is.
I used to think it was because there’s something inherently selfish in asking for something for free; that the very act of asking someone to work for free implies a certain disrespect for the value of their time. A client who is paying by the hour has a vested interest in not wasting the freelancer’s time, since they’re paying for it. Also, they will tend to value the finished piece more, as they can associate a dollar value to it.
That argument makes some sense, but I don’t think that’s it. If that were the case, the nature of the problem would be that pro bono clients tend to nitpick over the details of the work, asking for endless little tweaks and changes. That isn’t my experience, however. The problem I’ve run into again and again (four out of the last four times I’ve grudgingly agreed to do something for free, in fact), is that the work is actually rejected entirely – I’ve either been asked to try something completely different, make so many changes it would be easier to go back to the drawing board, or else the client has decided that they didn’t need the work, or didn’t like what I did, and that they would do it themselves, or ask someone else.
Of course, you could try to put the blame on the artist in cases like that, but the fact is that in five years of working as an artist/designer, I can count on one finger the number of times I’ve had a paying client express dissatisfaction with my work. Nor is it that I’m not putting as much effort into free pieces – I’ve put many hours into a pro bono piece only to see it discarded, unsued, and I’ve likewise taken on low-budget, quick-and-dirty paid work and received great praise and a happy client.
Rather, the conclusion I’ve reached is this:
Paying clients are going to pay for the work whoever they choose, so they seek you out because they like your work. They’ve checked out lots of portfolios, and they sought you out because your work has the look that they have in mind. Pro bono clients, on the other hand, seek you out because they think they can get something for free. Either they know you personally, or they know you’re interested in their cause, or for whatever other reason, they believe that you’re likely to agree to do something for them unpaid. They might not even have seen your portfolio, they just know that you’re in the business. The odds are therefore very slim that you also happen to be the person they would have chosen to do the work if they’d been choosing based on style. A mismatch between the artist’s style and the client’s vision never ends well.
Making lowball bids to win contracts has the same problem – if you bid slightly higher than the competition, you may get fewer contracts, but the clients who choose you will have done so because they want you, specifically. Bid too low, however, and not only will you fail to be compensated adequately for your work, you will also end up with clients who are secretly hoping that you’ll do something similar to what some other artist would have done, just more cheaply than him or her.
And so, while working cheaply or for free may seem like a nice thing to do, it may actually be a recipe for frustration and disappointment, both on your part and that of the person you’re working for. It’s important to examine the motivations of those who approach you to do work – the primary reason should always be that they like your work, and believe that your style will give them the product they’re looking for. If you believe their motivation might be anything else, think twice before accepting the job.